Andrew Song of Whisper AI on solving hearing loss with AI
The Robot Brains Podcast
00:00
The Import of a Speech Recognition Data Set
In the most basic construct of the model, what you're really trying to do is get a almost like an image mask. And that's where maybe we've innovated the most. We tend to r r data sets ar very speech focused, because most people, when they're in a noisy place, they want to hear the person talking to them. But that's not always true. Or you can imagine, if you're bird watching, you know, onle you care more about birds thand humans, right? All these different things. I am very excited about the kind of the longer range of this problem, why there's so much work to be done.
Transcript
Play full episode
